RandomXiong commited on
Commit
a97c711
·
verified ·
1 Parent(s): ffe3201

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -17,6 +17,9 @@ license: apache-2.0
17
 
18
  Quantized version of [deepseek-ai/Deepseek-V3-0324](https://huggingface.co/deepseek-ai/Deepseek-V3-0324)
19
 
 
 
 
20
 
21
  ### Model Optimizations
22
  These models were obtained by quantizing the weights and activations of DeepSeek models to mixed-precision data types (W4(int)A(FP)8 for MoE layers and FP8 for dense layers).
 
17
 
18
  Quantized version of [deepseek-ai/Deepseek-V3-0324](https://huggingface.co/deepseek-ai/Deepseek-V3-0324)
19
 
20
+ | Model| MMLU |
21
+ |-------|-------|
22
+ | novita/Deepseek-V3-0324-W4AFP8 | 0.8734 |
23
 
24
  ### Model Optimizations
25
  These models were obtained by quantizing the weights and activations of DeepSeek models to mixed-precision data types (W4(int)A(FP)8 for MoE layers and FP8 for dense layers).